ObjectiveThe widespread adoption of portable fundus cameras for primary care and community screening is hindered by limitations in current autofocus(AF) technologies. Image-based methods relying on sharpness evaluation require iterative searches, resulting in slow convergence, while projection-based techniques are susceptible to optical artifacts and calibration errors. To address these challenges, this study introduces a novel AF system based on direct wavefront sensing, designed to deliver simultaneous high speed, high precision, and operational robustness within the compact form factor essential for portable ophthalmic devices. MethodsOur approach fundamentally reimagines the AF process by directly measuring the ocular wavefront aberration. We developed a custom portable fundus camera integrating a miniaturized Shack-Hartmann wavefront sensor (SHWS) into the optical path. An 850 nm laser diode projects a point source onto the retina via oblique illumination to minimize corneal reflections. Light scattered from this spot carries the eye’s refractive error through the imaging optics and is directed to the SHWS, positioned at a plane optically conjugate to the primary color CMOS imaging sensor. A microlens array within the SHWS samples the incident wavefront, generating a pattern of focal spots on a CCD. Real-time centroid analysis of these spots provides a map of local wavefront slopes. These measurements are processed through a singular value decomposition (SVD) algorithm to fit a Zernike polynomial basis set, enabling real-time reconstruction of the wavefront phase. The defocus component (S) is extracted from the second-order Zernike coefficients, providing a direct, quantitative measure of the refractive error in diopters. This value serves as a precise error signal in a closed-loop control system, which commands a voice-coil actuated focusing lens to its null position in a single, deterministic step, eliminating the need for iterative search algorithms. ResultsComprehensive evaluation demonstrated the system’s high performance. Testing on a calibrated model eye (OEMI-7) established a highly linear relationship between the computed defocus S and the focusing lens position across a ±20 Diopter (D) compensation range, achievable within a 5 mm mechanical travel. The system achieved a focusing precision of 0.08 D, corresponding to an 18-fold improvement over a conventional projection spot-size method tested under identical conditions. The total focus acquisition time, encompassing wavefront measurement, computation, and lens actuation, averaged under 0.5 s. Clinical validation with 25 human volunteers (50 eyes, refractive range -15 D to +10 D) confirmed practical efficacy. The wavefront-sensing AF succeeded in 92% of attempts with a mean time of 0.5 s, substantially outperforming a projection-based benchmark which achieved only a 32% success rate with an average time of 4.25 s. The system provided instantaneous directional guidance and maintained stability during minor ocular movements. Objective assessment of image quality, via amplitude contrast of retinal vasculature, showed consistent and significant enhancement following AF correction across the entire tested diopter range. ConclusionThis work successfully implements and validates a direct wavefront-sensing autofocus paradigm for portable fundus cameras. By directly quantifying and compensating for the optical defocus aberration, this method bypasses the fundamental limitations of image-processing and projection-based techniques, enabling rapid, precise, and deterministic diopter compensation. The developed system delivers an exceptional combination of a wide operational range (±20 D), high accuracy (0.08 D), fast convergence (0.5 s), and a compact physical footprint. This technology provides a practical and high-performance focusing solution capable of enhancing the reliability, throughput, and diagnostic utility of portable retinal imaging in large-scale screening applications. Future efforts will be directed towards system cost optimization and performance adaptation for diverse ocular conditions.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveDiscriminating atypical hepatocellular carcinoma (HCC) from other malignancies in liver nodules classified as Liver Imaging Reporting and Data System category M (LR-M) remains a significant diagnostic challenge on conventional ultrasound examination. The LR-M category, originally intended to capture non-HCC malignancies, paradoxically contains up to 63% of atypical HCCs that deviate from classic enhancement patterns, leading to potential misdiagnosis and suboptimal treatment planning. While deep learning has shown promise in HCC diagnosis, most existing models rely exclusively on single-modality ultrasound, overlooking the diagnostic benefits of integrating complementary information from multiple imaging sources. To address this gap, we propose a novel attention-weighted tri-modal ultrasound network (TUS-Net) that integrates contrast-enhanced ultrasound (CEUS), B-mode ultrasound (BUS), and time-intensity curves (TICs) to improve diagnostic accuracy for these clinically challenging lesions. MethodsOur framework incorporates a three-dimensional convolutional neural network (C3D) backbone to extract spatiotemporal features from CEUS videos, capturing dynamic vascular patterns critical for lesion characterization. To effectively fuse complementary modalities, we introduce a dual-channel feature fusion module (DCFFM) that adaptively combines features from CEUS and BUS through channel-wise attention mechanisms, allowing the model to dynamically weigh the contribution of each modality based on diagnostic relevance. Additionally, we propose a temporal intensity feature fusion module (TIFFM) that leverages quantitative hemodynamic information from TICs to guide the model’s attention toward diagnostically critical temporal phases, such as arterial wash-in and portal venous washout. The model is further enhanced by automated lesion localization using YOLOX and class activation mapping for interpretability, ensuring that predictions align with clinically meaningful imaging features. ResultsEvaluated on a tri-modal ultrasound dataset comprising 161 patients with pathologically confirmed LR-M nodules (131 atypical HCC and 30 non-HCC malignancies), our model achieved an accuracy of 86.83%, a sensitivity of 92.50%, a specificity of 75.50%, and an AUC of 89.32% in screening atypical HCC. Compared to single-modality baselines, TUS-Net demonstrated superior specificity, a clinically critical metric given the higher risk associated with misclassifying non-HCC malignancies. Ablation studies confirmed the contribution of each module, with the full model outperforming both standard C3D and 3D ResNet backbones integrated with attention mechanisms. A reader study involving junior and senior radiologists further validated the clinical utility of AI assistance, showing consistent improvements in specificity and inter-reader consistency, particularly for less experienced clinicians. ConclusionThese results surpass existing benchmark models and demonstrate the potential of our approach to enhance diagnostic precision in clinically specific cases. By intelligently fusing multi-modal ultrasound data with attention-guided mechanisms, TUS-Net offers a reliable and interpretable tool that holds promise for improving the non-invasive diagnosis of atypical HCC in challenging LR-M liver nodules.

ObjectiveTo explore the clinical and phenomics characteristics of patients with phlegm-stasis binding syndrome and its accompanied patterns in stable angina pectoris （SAP） of coronary heart disease （CHD）. MethodsA multicenter cross-sectional study design was adopted. A total of 300 patients with SAP of CHD were enrolled and classified into 120 cases of phlegm-stasis binding syndrome， 125 cases of qi deficiency-accompanied syndrome， 38 cases of qi stagnation-accompanied syndrome， and 17 cases of toxin accumulation-accompanied syndrome according to traditional Chinese medicine （TCM） patterns. Data of patients with different TCM patterns were collected， including general condition， TCM symptoms， blood lipids， coagulation function， immune indicators， and serum metabolomics. Metabolic pathway enrichment analysis was used to compare differences in phenomics characteristics among groups. Metabolites with variable importance in projection （VIP） ≥1 and fold change （FC） ≥2 were considered as representative differential metabolites. ResultsPatients in each TCM pattern type were most commonly in the 60-75 years age group， with a relatively high proportion of males. Regarding TCM symptoms， patients with phlegm-stasis binding syndrome most commonly presented with wiry-choppy or wiry-slippery pulse， chest pain， and chest tightness； in patients with qi deficiency-accompanied syndrome， fatigue， chest pain， and weak pulse were most common； in patients with qi stagnation-accompanied sydnrome， wiry-choppy or wiry-slippery pulse， chest pain， and symptoms that increase or decrease with emotional changes， belching， or flatulence were most common； in patients with toxin accumulation-accompanied syndrome， bitter taste in the mouth， chest tightness， irritability， restlessness or manic delirium， and dry and hard stools or foul-smelling diarrhea were most common. Comparisons among the different TCM patterns showed statistically significant differences in coagulation parameters （P＜0.05）， whereas no statistically significant difference was found in blood lipid levels and immune indicators （P＞0.05）.Metabolomics analysis suggested that disorders of glycerophosphate metabolism and valine， leucine， and isoleucine metabolism are characteristic features of phlegm-stasis binding syndrome and its accompanied patterns. The representative differential metabolites between phlegm-stasis binding syndrome and qi deficiency-accompanied syndrome were 7,8-dihydrobiopterin （FC = 3.58） and oxypurinol （FC = 124.50）. Those between phlegm-stasis binding syndrome and qi stagnation-accompanied syndrome were cytidine 5′-diphosphocholine （FC = 2.62） and D-mannosamine （FC = 2.99）. Those between phlegm-stasis binding syndrome and toxin accumulation-accompanied syndrome were anserine （FC = 7.83） and canrenone （FC = 8.94）. ConclusionThere are certain differences in the clinical characteristics and phenomics characteristics among patients with SAP due to CHD exhibiting phlegm-stasis binding syndrome and its accompanied patterns. The phenomics characteristics mainly involve biological alterations such as lipid metabolism disorders， amino acid metabolism abnormalities， and multiple immune indicators activation， which may provide a reference for precise differentiation and treatment of SAP of CHD in TCM clinical practice.

ObjectiveTo explore the clinical and phenomics characteristics of patients with phlegm-stasis binding syndrome and its accompanied patterns in stable angina pectoris （SAP） of coronary heart disease （CHD）. MethodsA multicenter cross-sectional study design was adopted. A total of 300 patients with SAP of CHD were enrolled and classified into 120 cases of phlegm-stasis binding syndrome， 125 cases of qi deficiency-accompanied syndrome， 38 cases of qi stagnation-accompanied syndrome， and 17 cases of toxin accumulation-accompanied syndrome according to traditional Chinese medicine （TCM） patterns. Data of patients with different TCM patterns were collected， including general condition， TCM symptoms， blood lipids， coagulation function， immune indicators， and serum metabolomics. Metabolic pathway enrichment analysis was used to compare differences in phenomics characteristics among groups. Metabolites with variable importance in projection （VIP） ≥1 and fold change （FC） ≥2 were considered as representative differential metabolites. ResultsPatients in each TCM pattern type were most commonly in the 60-75 years age group， with a relatively high proportion of males. Regarding TCM symptoms， patients with phlegm-stasis binding syndrome most commonly presented with wiry-choppy or wiry-slippery pulse， chest pain， and chest tightness； in patients with qi deficiency-accompanied syndrome， fatigue， chest pain， and weak pulse were most common； in patients with qi stagnation-accompanied sydnrome， wiry-choppy or wiry-slippery pulse， chest pain， and symptoms that increase or decrease with emotional changes， belching， or flatulence were most common； in patients with toxin accumulation-accompanied syndrome， bitter taste in the mouth， chest tightness， irritability， restlessness or manic delirium， and dry and hard stools or foul-smelling diarrhea were most common. Comparisons among the different TCM patterns showed statistically significant differences in coagulation parameters （P＜0.05）， whereas no statistically significant difference was found in blood lipid levels and immune indicators （P＞0.05）.Metabolomics analysis suggested that disorders of glycerophosphate metabolism and valine， leucine， and isoleucine metabolism are characteristic features of phlegm-stasis binding syndrome and its accompanied patterns. The representative differential metabolites between phlegm-stasis binding syndrome and qi deficiency-accompanied syndrome were 7,8-dihydrobiopterin （FC = 3.58） and oxypurinol （FC = 124.50）. Those between phlegm-stasis binding syndrome and qi stagnation-accompanied syndrome were cytidine 5′-diphosphocholine （FC = 2.62） and D-mannosamine （FC = 2.99）. Those between phlegm-stasis binding syndrome and toxin accumulation-accompanied syndrome were anserine （FC = 7.83） and canrenone （FC = 8.94）. ConclusionThere are certain differences in the clinical characteristics and phenomics characteristics among patients with SAP due to CHD exhibiting phlegm-stasis binding syndrome and its accompanied patterns. The phenomics characteristics mainly involve biological alterations such as lipid metabolism disorders， amino acid metabolism abnormalities， and multiple immune indicators activation， which may provide a reference for precise differentiation and treatment of SAP of CHD in TCM clinical practice.

ObjectivePrimary liver cancer, predominantly hepatocellular carcinoma (HCC), is a significant global health issue, ranking as the sixth most diagnosed cancer and the third leading cause of cancer-related mortality. Accurate and early diagnosis of HCC is crucial for effective treatment, as HCC and non-HCC malignancies like intrahepatic cholangiocarcinoma (ICC) exhibit different prognoses and treatment responses. Traditional diagnostic methods, including liver biopsy and contrast-enhanced ultrasound (CEUS), face limitations in applicability and objectivity. The primary objective of this study was to develop an advanced, light-weighted classification network capable of distinguishing HCC from other non-HCC malignancies by leveraging the automatic analysis of brightness changes in CEUS images. The ultimate goal was to create a user-friendly and cost-efficient computer-aided diagnostic tool that could assist radiologists in making more accurate and efficient clinical decisions. MethodsThis retrospective study encompassed a total of 161 patients, comprising 131 diagnosed with HCC and 30 with non-HCC malignancies. To achieve accurate tumor detection, the YOLOX network was employed to identify the region of interest (ROI) on both B-mode ultrasound and CEUS images. A custom-developed algorithm was then utilized to extract brightness change curves from the tumor and adjacent liver parenchyma regions within the CEUS images. These curves provided critical data for the subsequent analysis and classification process. To analyze the extracted brightness change curves and classify the malignancies, we developed and compared several models. These included one-dimensional convolutional neural networks (1D-ResNet, 1D-ConvNeXt, and 1D-CNN), as well as traditional machine-learning methods such as support vector machine (SVM), ensemble learning (EL), k-nearest neighbor (KNN), and decision tree (DT). The diagnostic performance of each method in distinguishing HCC from non-HCC malignancies was rigorously evaluated using four key metrics: area under the receiver operating characteristic (AUC), accuracy (ACC), sensitivity (SE), and specificity (SP). ResultsThe evaluation of the machine-learning methods revealed AUC values of 0.70 for SVM, 0.56 for ensemble learning, 0.63 for KNN, and 0.72 for the decision tree. These results indicated moderate to fair performance in classifying the malignancies based on the brightness change curves. In contrast, the deep learning models demonstrated significantly higher AUCs, with 1D-ResNet achieving an AUC of 0.72, 1D-ConvNeXt reaching 0.82, and 1D-CNN obtaining the highest AUC of 0.84. Moreover, under the five-fold cross-validation scheme, the 1D-CNN model outperformed other models in both accuracy and specificity. Specifically, it achieved accuracy improvements of 3.8% to 10.0% and specificity enhancements of 6.6% to 43.3% over competing approaches. The superior performance of the 1D-CNN model highlighted its potential as a powerful tool for accurate classification. ConclusionThe 1D-CNN model proved to be the most effective in differentiating HCC from non-HCC malignancies, surpassing both traditional machine-learning methods and other deep learning models. This study successfully developed a user-friendly and cost-efficient computer-aided diagnostic solution that would significantly enhances radiologists’ diagnostic capabilities. By improving the accuracy and efficiency of clinical decision-making, this tool has the potential to positively impact patient care and outcomes. Future work may focus on further refining the model and exploring its integration with multimodal ultrasound data to maximize its accuracy and applicability.

Objective To explore the occurrence and development trajectories of symptoms at different time points in patients with acute pancreatitis (AP), and to analyze the influencing factors and preventive measures of development trajectories of AP symptom clusters. Methods A convenient sampling method was used to select AP who were admitted from January 2023 to December 2023 were selected and included in the study. The symptoms at different time points were recorded. The severities of symptom clusters in AP patients were explored, and the development trajectories of main symptom clusters were analyzed. Univariate and multivariate logistic regression analyses were used to analyze the influencing factors of development trajectories of symptom clusters in AP patients. Results The incidence rates of abdominal pain, dry mouth, abdominal distension and lack of energy were higher in AP patients during hospitalization. The incidence rates of lack of energy, anxiety, abdominal pain and sleep disturbance were higher on the 1^st month after discharge. The incidence rates of abdominal distension, abdominal pain, sleep disturbance and anxiety were higher on the 3^rd month after discharge. The incidence rates of anxiety, abdominal pain and irritability were higher on the 6^th month after discharge. The fatigue symptom cluster, psychological symptom cluster and gastrointestinal symptom cluster were extracted during hospitalization and on the 1^st month and the 3^rd month after discharge, and the psychological symptom cluster and gastrointestinal symptom cluster were extracted on the 6^th month. The severity scores of symptom clusters at each time point were statistically different (P<0.05). The development of gastrointestinal symptom cluster in AP patients was mainly low decline. The development of psychological symptom cluster was mainly high decline. Drinking history and diabetes mellitus were the influencing factors of development trajectory of gastrointestinal symptom cluster in AP patients (P<0.05). High disease severity, drinking history and biliary tract disease were the influencing factors of development trajectory of psychological symptom cluster in AP patients (P<0.05). Conclusion The symptom clusters of AP patients changes over time, with digestive, fatigue, and psychological symptoms being the main groups in the early stage, and psychological and digestive symptoms persisting in the later stage. Early identification and intervention are crucial for improving the prognosis of AP patients.

Intervertebral disc degeneration(IDD)is a prevalent clinical degenerative disease that currently can only be treated through conservative and surgical treatments,which only alleviate symptoms and are not significantly effective.In recent years,nanoparticles have been widely studied in the biomedical field due to their biodegradability,biocompatibility,extended body circulation,sustained and controlled release,and precise drug targeting.Nanoparticle drug delivery systems have the potential to deliver a range of therapeutic agents including proteins,drugs,genes,and cells,thereby promoting tissue and cell repair and regeneration,which offers hope for IDD treatment.However,there are still challenges in translating experimental data into practical therapies applicable to humans.This review summarizes recent research progress on drug delivery systems for IDD treatment based on nanoparticles and provides insights and prospects for the challenges faced by nanoparticles,aiming to provide a reference for the clinical translation of nanoparticle-based treatment for IDD.

Objective To investigate the effects of varying durations of overwork on cardiomyocyte pyroptosis in mice.Methods A total of 24 SPF KM mice were randomly divided into four groups(n=6)using a random number table:control group,2-week overwork(W2)group,4-week overwork(W4)group,and 6-week overwork(W6)group.Mice in control group were normally raised,while those in W2,W4,and W6 groups were forced to stand in water for 8 h and then restrained for 3 h daily for 2,4,6 weeks,respectively.The general condition and weekly weight changes of the mice were observed.After modeling,blood samples were collected,and hearts were excised.Myocardial histopathological changes were assessed using hematoxylin and eosin(HE)staining.The localization of gasdermin D(GSDMD)protein in myocardial tissue was detected through immunohistochemical staining,and the expression levels of pyroptosis-related proteins[NOD-like protein receptor 3(NLRP3),Caspase-1,GSDMD]in myocardial tissue were analyzed using Western blotting.The contents of interleukin-1β(IL-1β)and interleukin-18(IL-18)in serum and myocardial tissues were measured using ELISA.Results(1)The weight of control group mice increased steadily within 2 weeks.In W2 group,there was no significant weight change within 2 weeks,while in W4 and W6 groups,the body weights were higher than their initial values from the 2nd to 6th week.Compared with control group,the body weights of W2,W4,and W6 groups were lower than those of control group in the 1st and 2nd week,with statistically significant differences(P<0.05).The activity levels of the mice in W2,W4,and W6 groups initially increased and then decreased,with their fur becoming dull and falling out,and their mental state deteriorating.(2)In control group,cardiomyocytes were neatly arranged,and the nuclear morphology was normal.Compared with control group,in W2 group,cardiomyocyte arrangement was less regular,and capillary congestion was increased.In W4 group,the vascular congestion in the myocardium was significantly increased,the interstitial tissue was hyperplastic,and vacuolization appeared around the nuclei.In W6 group,the myocardial interstitium was loose,fat infiltration was increased,vacuolization around the nuclei was increased,and myocardial fibers were swollen,and the arrangement was disordered.(3)GSDMD was mainly located in the cytoplasm of cardiomyocytes.Compared with control group,the expression levels of NLRP3,Caspase-1,and GSDMD proteins in W2,W4,and W6 groups were significantly increased,and the expression levels were in the order of W6 group>W4 group>W2 group,with significant differences(P<0.05).(4)Compared to control group,the levels of IL-1β in serum and myocardial tissues of W2,W4,and W6 groups were significantly increased.In serum,the level of IL-1β in W6 group was higher than those in W2 and W4 groups,and in myocardial tissue,the levels in W4 and W6 groups were higher than those in the W2 group,with significant differences(P<0.05).There were no significant differences in IL-1β levels in serum among W2 and W4 groups,nor were there significant differences in myocardial tissue between W4 and W6 groups(P>0.05).Compared with control group,the levels of IL-18 in serum and myocardial tissue of W4 and W6 groups were significantly increased(P<0.05).In serum,the levels of IL-18 in W4 and W6 groups were higher than that in W2 group,and in myocardial tissue,the level in W6 group was higher than those in W2 and W4 groups,with the differences being statistically significant(P<0.05).Conclusions Overwork can cause structural damage to mouse myocardial tissue,increase the expression of pyroptosis proteins NLRP3,Caspase-1,GSDMD,and aggravate myocardial inflammatory responses in overworked mice.Cardiomyocyte pyroptosis may be one of the factors contributing to sudden cardiac death induced by overwork.

Objective:To analyze the pathological features of recurrent cardiac myxoma to provide a reference basis for clinical treatment and postoperative follow-up.Methods:The pathological data of cardiac myxoma patients who underwent cardiac myxoma surgery in Beijing Anzhen Hospital, Beijing, China from 2002 to 2016 were retrospectively analyzed. According to the grouping criteria, the cases were divided into the recurrence group ( n=6) and control group ( n=73). Results:In the recurrence group, there were 3 females and 3 males with a median age of 47 years. In the control group, there were 49 females and 24 males, with a median age of 53 years. Cardiac myxoma usually occurred in the left atrium. In the recurrence group, 5 cases occurred in the left atrium and 1 case in the right atrium. In the control group, 68 cases occurred in the left atrium, 4 cases in the right atrium, and 1 case in bilateral atria. Among the 6 cases in the recurrence group, the recurrence time was 1-7 years, with average 4.8 years. In the univariate analysis of recurrent cardiac myxoma pathology, disruption of elastic fiber layer and tumor cells of the tumor stalk invading into the myocardium through the elastic fiber layer were statistically associated with recurrence of cardiac myxoma ( P<0.05). Logistic regression analyses showed that the invasion of tumor cells through the elastic fiber layer of the heart wall into the myocardium was an independent risk factor for recurrence ( Odds Ratio=0.999, P<0.05). Conclusion:During the pathologic diagnosis, assessing the relationship between tumor cells in the stalk of cardiac myxoma and elastic fiber layer can help estimate the recurrence risk of cardiac myxoma, and thus guide clinical treatment and postoperative follow-up.